Disruptive Innovation in Stroke Care

Henry Feldman MD1, Stan Finkelstein MD2, Shane Reti QSM, MD1,

Steffen Bayer PhD3, James Barlow PhD3

1. Division of Clinical Informatics, Beth Israel Deaconess Medical Center, Harvard Medical School

2. MIT Systems Engineering Division

3. Imperial College Business School, London

Policy Context

US, UK and other developed health economies are

challenged to deliver high value healthcare and improve

clinical outcomes, while containing growth in costs

Stroke is one (of many) examples in which changes to the

care pathway for diagnosis and management are needed

to improve morbidity, mortality and long term costs

We will argue that an innovative field based imaging

device that would be widely available, could help

accomplish these aims

OECD Health Care Expenditure

0

2

4

6

8

10

12

14

16

181960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

OECD countries

Australia

Canada

France

Germany

Iceland

Ireland

Italy

Japan

Netherlands

New Zealand

Spain

Switzerland

United Kingdom

United States

Growth as percentage of GDP

Drivers of Growth in US Health Care Costs

Smith, Heffler and

Freeland (2000)

Cutler (1995) Newhouse (1992)

Aging of the Population 2 2 2

Changes in Third Party

Payment10 13 10

Personal Income Growth 11-18 5 <23

Prices in Health Care Sector 11-22 19 NA

Administrative Costs 3-10 13 NA

Defensive Medicine and

Supplier-Induced Demand0 NA 0

Technological Innovation 38-62 49 >65

PercentageSource: Peter Orszag, Congressional Budget Office, 2008

Innovation

In nearly every industry new product innovations initially are expensive, complicated, inaccessible and require skill (Christensen, 2008)

In most industries, technological innovation proves to be cost reducing as process innovation follows product innovation (Utterback, 1978)

Not observed in health care, as technological innovation has lead to cost inflation (Blume, 1992)

Doctors need to access the latest, most sophisticated technologies to meet what they perceive to be the needs of their patients (Fuchs, 1974)

Patients often consider availability of latest practices to be a measure of quality

A Way to View Failure of Innovation in

Health Care to be Cost-Reducing

“Quality Adjusted Price Indexes”

Initiated by Andrew Court, Automobile Industry Economist in

1939

Popularized by Leading Academic Economists from 1970 to

Present

Concept is that each time automakers introduce new

models, they are offered at higher prices, but also have new and

innovative characteristics and attributes

Quantitative approach is to assign value to these attributes and

adjust growth in prices accordingly

A Way to View Failure of Innovation in

Health Care to be Cost-Reducing (cont)

The result for industry sectors, including

automobiles, personal computer, consumer

electronics, etc, is that adjusted growth rates in prices are

highly negative

Similar approach applied to health care

(pharmaceuticals, insurance premiums, etc) finds some

moderation in growth rates compared to unadjusted

price increases, but not nearly as dramatic as for many

other industries

Disruptive Technology

Once in a while a new technology comes along with

these characteristics (Christensen)

It is simplifying compared to current practices

It is less costly

Initially not as good

As the technology improves, it disrupts the standard of

practice

Disruptive Technology: Steel IndustrySt

eel Q

ual

ity

1975 1980 1985 1990

Rebar

Angle Iron, bars and Rods

Structural Steel

Sheet Steel

7%

12%

12%

25-30%

Source: Christensen

Even Disruptive Technology in Health Care

Doesn’t Always Produce the Desired Effect

Treatment of Coronary Artery Disease

Old Standard of Practice: Coronary Artery Bypass Graft (CABG)

Professional Group: Cardiac Surgeons

Cost: $170,000 for surgery, pre-op, post-op and follow up care

Disruptive Technology: Balloon Angioplasty with Stents

Professional Group: Interventional Cardiologists

Cost: $29,000

Impact of Disruptive Technology: Huge growth in volume of patients receiving

treatment. Health care costs rise greatly rather than fall

Stroke Care: Proposal for Innovative Field

Based Imaging Device

Stroke care does not require high-resolution capability of

stationary hospital based CT-scanning units

Significantly cheaper

Require less space

Mobile

Uses a less skilled operator

Economy of scale for low incidence per operator use of

expensive consultant to read result

Time to treatment reduced

Stroke Background

UK has about 150,000 strokes per year, US has around

780,000

One type of stroke can be treated with “clot busting

medications” such as tPA

At our last presentation we discussed that while in the

US almost all eligible stroke patients were treated with

tPA and in the UK, the majority were not, this has

changed in major metropolitan areas with installation of

stroke centers

In rural areas however, strokes are still not routinely treated

with tPA due to lack of advanced imaging

Stroke Types

By definition, strokes are due to a disruption in blood flow to an area of the brain, there are however 2 causes of this which necessitate major changes in therapy:

Ischemic Stroke: This is due to a vessel being blocked by either clot or a combination of clot and cholesterol plaque

This requires use of a clot busting agent (thrombolytic)

You have a small number of hours (the window) to administer

Hemorrhagic Stroke: This is due to a blood vessel tearing and bleeding into the brain.

Clot busting agents will worsen this stroke

The key is not diagnosing but rather determining the type of stroke, as this changes management

Stroke Diagnosis Technology

Current state of the art is use of a non-contrast agent

based CT-scan of the brain. This shows bleeding

Why This Approach Is Ripe For Disruption

CT-Scanners are very expensive ($150,000-$1m) and

require specialized facilities and operators

In the US they are very common due to reimbursement

strategy, but in the UK they are much less so

The output of a CT-scan is currently interpreted by a

radiologist typically at the site of the CT-scan

In a busy center this makes sense, but in low incidence areas

such as rural areas, this is very cost inefficient

In rural areas the ability to bring a patient to the CT-scan

in time, while their brain is dying, is problematic

There are “portable” CT-scanners but they are still quite

expensive and would take up most of an ambulance

Disruption in Stroke Care Proposal

The UK has already demonstrated that field based administration of tPA is safe and effective in MI (Heart Attack) therapy by lower-level providers (Paramedics)

The ability to give tPA or not in an MI, is based purely on background/history of the patient, and does not require additional imaging (other than the EKG)

In stroke care you need an image of the brain to determine the risk of worsening the stroke with tPA

Note: in New Zealand they do not, and accept the 6% mortality from this

We propose to move imaging to the field with simplified technology using, less skilled operators in low incidence areas

Ultrasound Technology

Ultrasound has been used to image the brain, in

babies, however in adults the solid skull prevents good

imaging

The skull radiates sound in all directions like a spherical

speaker

The skull distorts sound in and out variably based on the

location on the skull

The thinnest portion of the skull is at the temporal bone

Ultrasound devices are currently very inexpensive

($10,000) and are the size/weight of an iPad

Field ruggedized versions exist already, but there is no sensor

on the market that can image the brain of an adult

Acoustic Coupling Problem

Noise Cancellation

Ultrasound Helmet

Disruptive Additions

Computer power is so inexpensive and available, that

digital signal processing to measure and compensate for

the skull distortion is now available at low cost

Noise cancellation technology is now extremely

inexpensive and available as single chips

Ability to send images around the world, even via 3G is

now a standard part of all imaging devices

Why This Makes a Difference

With a direct imaging technology on the

ambulance, remote reading by a centralized

radiologist, we can directly give tPA in the field with rapid

treatment, long term disability is reduced

Modern 3G networks allow direct field support by

advanced providers anywhere in the world

Herd Immunity Function

Early modeling work, seems to suggest that you don’t

even need to equip every ambulance with the technology

to meet the full demand.

Micro-regional distribution would allow deployment in low

incidence areas rapidly

Ongoing Questions

Would the NHS (or another health care system) be able

to make the organizational change required to completely

change the stroke model from hospital to pre-hospital

based care?

Would this approach improve clinical outcomes?

Would this reduce costs? What changes to payment and

funding rules would be required?

Would this approach have broader applicability than just

stroke?

Conclusion

Disruptive technologies offer the potential to address the

widely observed phenomenon that technological

innovation in health care – differently from in other

industries – tends to be cost increasing

This effect is the opposite of what is commonly seen in

other industries

Field-based imaging using ultrasound promises to

substantially improve stroke care by enabling the more

widespread use of tPA

While the technological hurdles are high, the process

innovation hurdles are even higher

Road Sign